Construção de região de segurança em regime permanente para sistemas de distribuição de energia utilizando otimização por enxame de partículas

Abstract

The main focus of this dissertation is the study of steady-state security regions (SSR) for power distribution systems. This involves evaluating the operating points at which the system is capable of operating within the limits of the equipments and the criteria defined for the quality of the power supply. Initially, a bibliographical review is carried out, highlighting the evolution of the method, which originated in transmission systems, up to its recent and incipient application in distribution systems. The relevance of SSR for studies associated with the operation of electrical systems and grid expansion planning is emphasized. This tool is widely used within the area of electrical power systems (EPS) but is unusual and limited in extent in distribution systems. However, with the recent growth of distributed energy resources (DERs), tools that assess operational security have become essential to enable the energy transition process. This is followed by a review of the power flow solution process for distribution systems, the main input used to build the SSR. With the power flow background, the algorithm traditionally used in EPS to build the SSR is detailed, consisting of solving successive power flows from a given converged operating point, subject to fixed power increments. The method is developed for power distribution systems by integrating it with OpenDSS, which is used as a calculation engine to solve the power flow. Since the traditional methodology is computationally expensive, a new method for constructing the security region is proposed, based on the particle swarm optimization (PSO) process, with the aim of reducing the processing time of construction of the security region since it uses a variable power increment throughout the iterative method. With this same objective, it is detailed how the use of parallel processing can help in the construction of the SSR and how it can be used in the aforementioned methods. With both methodologies in place, security regions are constructed for two distribution systems, and the results obtained show that the proposed methodology is effective in constructing the SSR, in some cases showing better quality in defining the shape of the safe region. However, its application proved to be inefficient, since the processing time was generally longer than the traditional method, restricting the use of this method to specific cases. Finally, an application made possible by the use of SSR is presented, in which it is evaluated how the participation of distributed generation (DG) in local voltage control is able to increase the system’s safety margin, optimizing grid operation.